Forecasting Stock Market Indices with the Composite Leading Indicators: Evidence from Turkey

Abstract

The objective of this study is to evaluate the empirical performance of composite leading indicators (CLIs) in forecasting stock market indices for Turkey in the period from 2007:03 through 2019:07. After examining the stationarity of the series by using Narayan and Popp (2010) and Enders and Lee (2012) Fourier ADF unit root tests, the causality relationship from the composite leading indicators to stock market indices are tested by employing Enders and Jones (2016) Fourier Granger causality test. The results support the evidence of a causality relationship from composite leading indicators to BIST100, BIST Financial and BIST Industrial Indexes under structural breaks.